A novel band selection architecture to propose a built-up index for hyperspectral sensor PRISMA

Processing of hyperspectral remote sensing datasets poses challenges in terms computational expense pertaining to data redundancy. As such, band selection becomes indispensable address redundancy while preserving the optimal spectral information. This paper proposes a novel architecture using Genetic Algorithm (GA) optimizing technique with Random Forest (RF) classifier for efficient Hyperspectral Precursor Application Mission (PRISMA) dataset. The bands are BLUE (λ = 492.69 nm), NIR 959.52 and SWIR 1 1626.78 nm). also involves an application selected accurately identify quantify built-up pixels by means new index named Imagery-based Built-up Index (HIBI). proposed was used map six cities around world namely Jaipur, Varanasi, Delhi, Tokyo, Moscow Jakarta establish its robustness. analysis shows that has accuracy 94.02%, higher than all other indices considered this study. Moreover, separability establishes efficiency differentiate from spectrally similar land use or cover classes.